Dall-e 3 Announcement Breakdown

September 22, 2023

Tory Barber

Dall-e 3 Logo on a space background

OpenAI just announced it will soon be releasing 3rd model of Dall-e, appropriately named Dall-e 3.

Here are the highlights of this announcement.

When Does Dall-e 3 Release?

OpenAI is releasing Dall-e 3 to ChatGPT Plus and ChatGPT Enterprise users sometime in October.

It will then be released to API and Lab* users later in the fall.

*OpenAI Labs is the name given to the public online Dall-e image generator.

It is unclear when (or if) it will be integrated into ChatGPT for free users.

No exact dates have been given for either release yet.

What Makes Dall-e 3 Better?

Dall-e 3 ChatGPT integration

The biggest and most exciting change coming with the Dall-e 3 update is that the image generator will be directly integrated with ChatGPT!

You can seamlessly converse with the AI across multiple topics, brainstorming prompt ideas, and generating/iterating on images as you go.

OpenAI showcased an example where a parent generated an image of their child’s imaginary friend, built a persona around the character, and then asked the AI to write and illustrate a children’s story about the character they had built in that conversation. 🤯

Dall-e 3 is Just… Well, Better

Reducing the barrier to entry was a big theme of this announcement. Dall-e 2 is arguably one of the worst image generators out on the market right now. An in-depth knowledge of prompt engineering is required to get passable images.

Largely, Dall-e 3 is supposed to be better at being more accessible to a less technical audience. It is supposed to understand people’s prompts more.

Of course, we will see how it performs once it’s released, but if OpenAI’s examples can be believed, the updated Dall-e model looks leaps and bounds better than its predecessor. Just look at the difference between these two images generated from each model using the same prompt!

Dall-e 2

Dall-e 2 image of a basketball

Dall-e 3

Dall-e 3 image of a basketball

Prompt: “An expressive oil painting of a basketball player dunking, depicted as an explosion of a nebula.”

Bigger Focus on Safety, Privacy, and Artistic Ownership

The last big announcement is that this latest version of Dall-e has better systems in place to safeguard its users from the less desirable parts of AI image generations.

Namely, it blocks requests to generate:

  • images of public figures
  • propaganda or misinformation
  • Nude or explicit imagery
  • Harmful biased images

OpenAI worked with red teamers across the world to stress test this new system in order to make the platform as safe as possible.

Dall-e 3 will also give a little power back to artists. Artists will now be able to opt out of letting Open AI train its next model on their images. Additionally, Dall-e 3 will ignore the names of living artists in prompts. Meaning, it won’t be possible to mimic the work of living artists who are trying to make a living with their art.

Bigger Focus on Safety, Privacy, and Artistic Ownership

The last big announcement is that this latest version of Dall-e has better systems in place to safeguard its users from the less desirable parts of AI image generations.

Namely, it blocks requests to generate:

  • images of public figures
  • propaganda or misinformation
  • Nude or explicit imagery
  • Harmful biased images

OpenAI worked with red teamers across the world to stress test this new system in order to make the platform as safe as possible.

Dall-e 3 will also give a little power back to artists. Artists will now be able to opt out of letting Open AI train its next model on their images. Additionally, Dall-e 3 will ignore the names of living artists in prompts. Meaning, it won’t be possible to mimic the work of living artists who are trying to make a living with their art.

Thats it! Pretty short, but impactful announcement! We’ll see how the model actually performs in October, but for now I’m pretty excited about Dall-e 3!

Tory Barber Headshot Thumbnail

About Me

Hi I’m Tory!I’m a Graphic Designer, Web Developer, and AI OBSESSED artist from good ol’ Canada, eh!I record all the things I learn as I dive deeper into the Generative AI rabbit hole so I can teach people like you!

Follow Me

Prompt Inspo In Your Inbox!

Latest Articles

Lighting Prompts for Midjourney

Lighting Prompts for Midjourney

Dive into the world of Midjourney photography & understand the transformative influence of lighting. Enhance your visuals and create stunning portraits with these tested keywords.

Latest Posts

Lighting Prompts for Midjourney

Lighting Prompts for Midjourney

Dive into the world of Midjourney photography & understand the transformative influence of lighting. Enhance your visuals and create stunning portraits with these tested keywords.